Preferential secretion of inducible HSP70 by vitiligo melanocytes under stress

Abstract

Inducible HSP70 (HSP70i) chaperones peptides from stressed cells, protecting them from apoptosis. Upon extracellular release, HSP70i serves an adjuvant function, enhancing immune responses to bound peptides. We questioned whether HSP70i differentially protects control and vitiligo melanocytes from stress and subsequent immune responses. We compared expression of HSP70i in skin samples, evaluated the viability of primary vitiligo and control melanocytes exposed to bleaching phenols, and measured secreted HSP70i. We determined whether HSP70i traffics to melanosomes to contact immunogenic proteins by cell fractionation, western blotting, electron microscopy, and confocal microscopy. Viability of vitiligo and control melanocytes was equally affected under stress. However, vitiligo melanocytes secreted increased amounts of HSP70i in response to MBEH, corroborating with aberrant HSP70i expression in patient skin. Intracellular HSP70i colocalized with melanosomes, and more so in response to MBEH in vitiligo melanocytes. Thus, whereas either agent is cytotoxic to melanocytes, MBEH preferentially induces immune responses to melanocytes.

Keywords: autoimmune; heat shock protein; melanocyte; stress; vitiligo.

Figure 1

HSP70i overexpression in vitiligo skin. (A) Representative immunoperoxidase staining of HSP70i in vitiligo skin displaying expression predominantly located in the epidermis, with minimal cellular expression observed in nonlesional (left), and moderate to strong expression in perilesional (middle) and lesional (right) skin. (B) Subjective, blinded quantification of HSP70i expression in tissue sections. Immunoperoxidase intensity: 1=low, 2=medium, 3=high. Data are presented as mean ± SEM as calculated by Student's t-test. *P< 0.05, **P<0.01, n=7.

Figure 2

Cell viability of vitiliginous melanocytes. Primary melanocytes from control and vitiligo patients were plated at 10,000 cells per well in triplicate, and exposed to 125, 250, or 500 μM concentration of 4-tertiary butyl phenol (4-TBP) or monobenzyl ether of hydroquinone (MBEH) for 72 hours. Percentage viability was quantified in MTT assays, and compared relative to vehicle treated controls. Cell viability decreased as 4-TBP and MBEH dosage increased; however, no differences were observed between control and vitiligo melanocytes for any treatment. Data provided from two independent experiments with triplicate values for each experiment. Healthy control (n =8), vitiligo (n = 5). Data are presented as mean ± SEM as calculated by Student's t-test.

Figure 3

HSP70i colocalizes with melanosomal fractions. (A) Adult melanocytes (HM162P7) were dounced and underloaded on an iodixanol gradient for ultracentrifugation. Dense bands containing melanin were observed in the melanosomal (20/21 and 23) fractions. (B, upper image) Collected fractions (non-melanosomal 7–9, and melanosomal 20–22) were fixed and analyzed by electron microscopy (EM). (B, lower image) Collected fractions were run on a Western Blot. SPA-820 antibody (which binds both constitutive and inducible HSP70 isoforms) reacted with the melanosomal (19–24) but not nonmelanosomal fractions (1–15). (C) Quantification of western blot band intensities shown in EM. Mean luminosities indicate stronger antibody reactivity in fractions 19–24 versus 1–18. (D) Western blot analysis of non-melanosomal (1, 2 and 18) and melanosomal (19–23) fractions probed with the antibody SPA-810 which only detects inducible HSP70 (70 kD), and HMB45 to the melanocyte antigen gp100 (HMB45 detects a 45 kD product of gp100).Only the melanosomal fractions (19–23) react with both HSP70i and gp100 antibodies. Purified HSP70i protein was probed as a control. Together these data indicate that HSP70i is localized within the melanosome containing fractions.

Figure 4

HSP70i is overexpressed in vitiliginous melanosomes after stress. (A) Representative serial 1.0 μM Z-sliced images of neonatal melanocytes (Mf0627P11) indicate cytoplasmic expression of HSP70i (SPA-811 Ab detected by FITC) and the melanocyte antigen TRP-1 (Ta99 Ab detected by PE/Cy7) throughout the cell, but not present in the nucleus (DAPI counterstained). (B) Representative 0.5 μM Z-slice images of neonatal melanocytes (Mf0627P11) probed with antibodies to HSP70i (SPA-811) and gp100 (HMB45). Individual channels for HSP70i (GFP), gp100 (PE/Cy7) and merged images are shown. Note extracellular detection of HSP70i/gp100. An image pseudocolored by the ImageJ plug-in JACoP indicates levels of HSP70i/gp100 colocalization as low (blue), middle (yellow) and high (red). Perinuclear red mapping indicates over-lapping between the two labels, suggestive of colocalization. (C) Vitiligo and neonatal melanocytes were treated with 250 μM bleaching agents (4-TBP and MBEH) for 24 hours followed by confocal microscopy. Five z-slices from representative treated melanocytes were analyzed for HSP70i/gp100 colocalization using JACoP, with an nMDP cutoff of 1 used in the calculations. Data is presented as HSP70i/gp100 colocalization relative to total HSP70i staining. Graphed is the normalized mean deviation production (nMDP) for 5, 1μm Z-slices for each sample. Healthy control (n =3), vitiligo (n = 3). Data are presented as mean ± SEM as calculated by Student's t-test.* P<0.05, ***P< 0.001

Figure 5

Transmission electron microscope detection of HSP70i in 4-TBP treated melanocytes. Healthy melanocytes (Mf0632P2) were treated with 125 μM 4-TBP for 4 hours followed by immunostaining with gold particle-conjugated antibodies to HSP70i. The melanocytes were next fixed and scanned by transmission electron microscopy. Gold particles (red arrows) are observed in sections probed with anti-HSP70i which are more abundant (P < 0.0002) in 4-TBP treated melanocytes as shown in the bar graph below. HSP70i is seen throughout the cytoplasm in line with data shown in Figure 4, and occasionally juxtapositioned to melanosomes (insets). Bar equals 500 nm. (inset bars equal 40nm).

Figure 6

Vitiligo melanocytes secrete more HSP70i in response to MBEH. (A) The immortalized melanocyte lines (vitiligo PIG3VP54 and control PIG1P96; n=3 measurements) and (B) primary healthy and vitiligo melanocytes were treated with 125 um MBEH for 24 hours. Supernatants from treated and untreated cultures were assessed for HSP70i content by high-sensitivity ELISA. The percent increase in HSP70i of MBEH and vehicle treated compared to untreated samples is indicated. Images of cell cultures (A and B) show no differences in cell density or death after MBEH treatment. A significant increase in HSP70i secretion was observed in vitiligo, but not control melanocytes after MBEH treatment. These data indicate that melanocytes obtained from vitiligo skin secrete more HSP70i in response to stress than healthy cells. Control (n =4 individual cultures), vitiligo (n = 3 individual cultures). Data are presented as mean ± SEM as calculated by Student's t-test. **P< 0.01, ***P< 0.001.